Detection And Recognition Of Catenary String And Pantograph Skateboard Based On Image Processing Technology

In recent years,with the increasing volume of railway operations and the community’s emphasis on energy efficiency,electrified railway with its traction power,energy saving and environmental protection,can greatly improve the transport capacity and speed,and has technical,economic and environmental advantages,Has become a national railway priority development of the railway traction power.The rapid development of China’s high-speed railway in recent years has also made the electrified railway in the existing operation of all the proportion of the railway is growing.In the context of the rapid increase in the railway,the safety of traffic has been paid more and more attention,becoming an important issue of railway operation.And rely on manpower to carry out the traditional way of inspection is not only inefficient,high cost,but also have a long period to detect shortcomings,can not meet the requirements of the people,so the use of image processing technology Intelligent inspection method more and more people’s attention.The research work of this paper is based on the technical specifications of the pantograph safety inspection device and the catenary safety inspection device in the 6C system.The algorithm is based on the data of the pantograph and the catenary inspection,Using the technical means of picture processing to realize the intelligent identification of the sensitive equipment in the inspection picture,and also demonstrate the effectiveness of the proposed algorithm by experimental test.The main work and innovation of this paper include the following aspects:In the process of catenary sine detection,the Ridge Filter based on Hessian matrix is used to filter the picture,and then the filtered image is subjected to Huffman linear detection.All straight lines are screened by setting the threshold method,exclude the line does not meet the conditions,then the remaining line is the line on behalf of the hanging string,to achieve the exact identification of the catenary hanging string.In the pantograph skateboard detection,the method of flip detection is adopted.The method effectively increases the number of training samples in the case of less experimental data,which makes the classifier get more effective training and improve the accuracy of the program detectionIn the detection of pantograph skateboard,the HOG feature and the generalized Hoffmann transform are used to identify the pantograph skateboard.The method not only effectively removes the interference of the complex steel frame structure,but also more ideal to improve the detection efficiency of the program.Finally,the experimental test is used to collect the pantograph image and the catenary inspection picture.The algorithm has good applicability,and the ideal recognition rate is obtained.The two methods proposed in this paper are verified having a certain effectiveness.